Test results on WikiTree

[Chris Whitten]

Thanks to Bill and Val for opening the discussion here last week.

I'm really at a loss for how WikiTree can attach DNA test results to
profiles, or attach profiles to DNA test results.

For starters, I'm grossly ignorant of the subject. I had a 12-marker
test done through Family Tree DNA a few years ago but didn't find much
interesting in the results. A lot of the science went over my head. I
suppose I should get a better test done and really dig in to what it
means. Then maybe I could participate in this discussion more
confidently.

For those of you who have used DNA testing, have you tried integrating
results on WikiTree yet?

The free wiki bio space on a profile can include almost anything. But
it's information attached to an individual.

Another way to approach it would be to use categories. With
categories, individuals can be grouped in almost any way imaginable.

Any more input? Maybe the next step is for me to upgrade my test
results. Val, what would you recommend I get?

Thanks!

Chris

--
Chris Whitten
Creator of WikiTree.com
http://www.wikitree.com/wiki/Whitten-1

[Bill Atkinson]

I am one step back from you Chris, I have never had a DNA test done
but that was before I started my one name study and started to get
some idea of how they were using DNA in the Guild. I will do a bit of
study on their set of articles that have been done recently and give
you an update in the next week or so.

[Peter G]

I had a Y-DNA 67 test from Family Tree DNA a few years ago. The last name results matched what I expected, but also provided a goal for paper tracing. Current paper status only goes back to about 1900, but the DNA has a 66 out of 67 match to somebody who has a paper trace & same last name to 1595.

 

I don't know enough about Y-DNA to set up what the catagories should be. Off the top of my head, I'd have a Y-DNA start page with a link to family projects page(s); plus marker hierarchy. A person would scroll down the marker hierarchy to find their 12 marker (? maybe less) set and then branch off from there to seperate pages for each group.

 

End result would be people attached to the various levels of testing. So Chris, you'd be linked to your 12 marker results by name as actually tested; plus your male line would also be linked as a last name on that page maybe. Problem would be the size of the pages, so my idea needs work.

 

Peter 

---

From: Chris Whitten cwhi...@gmail.com

[William L Smith]

While I believe the discussion is useful, my first and second reaction is that, at this time, at least, it is an unnecessary and undesirable distraction from the important work that WikiTree is set up to accomplish. There are other resources available for DNA studies, where people are already knowledgeable.
Thank you for allowing me to insert my two cents worth. Just one opinion.

Dr. Bill  ;-)

--
You received this message because you are subscribed to the Google
Groups "DNA and Wiki Genealogy" group associated with http://www.WikiTree.com
To post to this group, send email to
dna-and-wik...@googlegroups.com
To unsubscribe from this group, send email to
dna-and-wiki-gene...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/dna-and-wiki-genealogy?hl=en

--
William L. Smith, Ph. D
billsm...@gmail.com
http://thehomeplaceseries.blogspot.com/
https://www.createspace.com/3441320
http://stores.lulu.com/drbillshares
http://drbilltellsancestorstories.blogspot.com/
http://thekinnickproject.blogspot.com/
http://eduscapes.com/v2a/back_to_the_homeplace.html
http://www.examiner.com/x-53135-Springfield-Genealogy-Examiner
http://www.examiner.com/x-58285-Ozarks-Cultural-Heritage-Examiner
http://drbillonretirement.blogspot.com/

[Hayduke]

Another way to provide DNA information is through links to the various DNA surname projects. My DNA information is included in the Lewis Surname DNA project at http://lewissurnamednaproject.com/ There are DNA charts on this site that compare the DNA profiles of members of the project that can be compared with those of WikiTree members. I tried putting my DNA profile on my Profile Page, but the size constraint for comments or memories doesn't allow the formatting necessary to show the full profile in a usable way. If we could have a box on our profile page that stretched the full width of the page, we could put our DNA profile there.

Michael

--
You received this message because you are subscribed to the Google
Groups "DNA and Wiki Genealogy" group associated with http://www.WikiTree.com
To post to this group, send email to
dna-and-wik...@googlegroups.com
To unsubscribe from this group, send email to
dna-and-wiki-gene...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/dna-and-wiki-genealogy?hl=en

--
Hayduke Blogs
http://hayduke2000.blogspot.com

[Val Van Zee]

I don't see how it would be productive to attach DNA results in individual
WikiTree profiles. And I don't understand enough about WikiTree to envision
how categorization might work or what it might look like. That said, here
are a few comments I hope you will find helpful.

The result numbers -in themselves- mean nothing... they are only valuable
when compared to matching results with the same surname. The exception to
the single surname rule is that it could be helpful in proving unrecorded
adoptions... where two or more of those tested to at least 67 markers show a
high probability of relationship. The best I can envision is -as far as
y-dna surname studies are concerned- is to sight the results page on the
corresponding surname project website... augmented with explanation and
discussion in simple layman's language . At it's best, DNA is just one
source of "proof" and we need a well researched line with lots of "proof"
sighted to make a sound analysis... coming in the form of traditional
research. There may be something I missing here, so as you try to think
about how WikiTree might best incorporate all this, think "matches" not
strings of result numbers.

Chris, you are not alone! I don't I understand the science part either. My
eyes go bleary just looking at those blown up segments of DNA strands and
molecules marked with the four little letters A, C, G and T. I don't
understand how anyone can make any sense of all that. Nevertheless, in
application it works a great deal of the time. In the large majority of
cases 12 marker tests are pretty useless. It will give you your projected
haplogroup so for someone who just wants that much can find it for 100
bucks. There are exception, but most 12 marker results will have many, many
matches. Hopefully at least some of these will be the expected surname, but
it is not unusual to have literally hundreds of matches -at that level- with
many different surnames, especially if you find yourself in a common
haplogroup such as R1b. With the y-dna database expansion over the last few
years, this phenomenon is not unusual even at the 25 marker level, though
instead of hundreds of matches this will usually be reduced to just a few to
maybe 100 other surname matches. A close match at these levels does
indicate that you share a common ancestor, but that common ancestor may have
lived many hundreds (or even many thousands) of years ago, often before the
common usage of surnames.

At 37 markers you begin to get more meaningful matches. When you have
matches at that level the chances are high that you share a common ancestor
within a genealogical time frame. With few exceptions, one should not
consider testing these days on less than 37 markers. IF there are matches
here, you need to get busy comparing paper trails, meanwhile waiting for
your results to come in for an upgrade to 67 marker.

I can foresee that future genealogies will not be considered "complete"
without some DNA results. Just as if we saw a genealogy today that left out
some important segment of documentation, like no census analysis.

Val

--------------------------------------------------
From: "Chris Whitten" <cwhi...@gmail.com>
Sent: Tuesday, May 31, 2011 9:11 AM
To: <dna-and-wik...@googlegroups.com>
Subject: [Wiki DNA] Test results on WikiTree

[Hayduke]

In my case, I have discovered unknown "cousins" who match my DNA profile but have no documented genealogical connection to my line. I'm aware of a Lewis line in the east coast of the US that do not know their genealogical heritage, but match my DNA. So I know they are related at some point. This gives them (and me) a place to start looking for documentation heretofore unknown.

Hence the DNA profile on our Profile page, or, at the least, a link to a web site that contains our DNA information.

It's the combination of DNA results and documentation that provides the full picture, eliminating unknown adoptions or clandestine visits by the ice man.

Michael

On Tue, May 31, 2011 at 1:56 PM, Val Van Zee <va...@bellsouth.net> wrote:

I don't see how  it would be productive to attach DNA results in individual WikiTree profiles. And I don't understand enough about WikiTree to envision how categorization might work or what it might look like. That said, here are a few comments I hope you will find helpful.

The result numbers -in themselves- mean nothing... they are only valuable when compared to matching results with the same surname. The exception to the single surname rule is that it could be helpful in proving unrecorded adoptions... where two or more of those tested to at least 67 markers show a high probability of relationship. The best I can envision is -as far as y-dna surname studies are concerned- is to sight the results page on the corresponding surname project website... augmented with explanation and discussion in simple layman's language . At it's best, DNA is just one source of "proof" and we need a well researched line with lots of "proof" sighted to make a sound analysis... coming in the form of traditional research. There may be something I missing here, so as you try to think about how WikiTree might best incorporate all this, think "matches" not strings of result numbers.

Chris, you are not alone! I don't I understand the science part either. My eyes go bleary just looking at those blown up segments of DNA strands and molecules marked with the four little letters A, C, G and T. I don't understand how anyone can make any sense of all that.  Nevertheless, in application it works a great deal of the time. In the large majority of cases 12 marker tests are pretty useless. It will give you your projected haplogroup so for someone who just wants that much can find it for 100 bucks. There are exception, but most 12 marker results will have many, many matches. Hopefully at least some of these will be the expected surname, but it is not unusual to have literally hundreds of matches -at that level- with many different surnames, especially if you find yourself in a common haplogroup such as R1b. With the y-dna database expansion over the last few years, this phenomenon is not unusual even at the 25 marker level, though instead of hundreds of matches this will usually be reduced to just a few to maybe 100 other surname matches.  A close match at these levels does indicate that you share a common ancestor, but that common ancestor may have lived many hundreds (or even many thousands) of years ago, often before the common usage of surnames.

At 37 markers you begin to get more meaningful matches. When you have matches at that level the chances are high that you share a common ancestor within a genealogical time frame. With few exceptions, one should not consider testing these days on less than 37 markers. IF there are matches here, you need to get busy comparing paper trails, meanwhile waiting for your results to come in for an upgrade to 67 marker.

I can foresee that future genealogies will not be considered "complete" without some DNA results. Just as if we saw a genealogy today that left out some important segment of documentation, like no census analysis.

Val

[John Hunter]

Let me say from the outset that I am a big fan of genetic testing as a
genealogy tool. I belong to a surname group that has had great success
augmenting their already extensive paper record. Also anything less
than a 67 gene test is not really worth all that much as the test
lacks definition at lower levels. If we tested at the one gene level
we would most probably match a good proportion of the world's male
population.

Basically if you match, or closely match, someone (irrespective of
their surname) on the 67 gene test then you know with great certainty
that you share a common ancestor somewhere in the past 300 or so
years. I have seen many paper record blocks removed with this tool as
has been my own experience. Once that match has been established it
helps to focus the record search that is then needed to get the paper
record of the relationship.

How the technology works - who cares, it works. The "male line"
determiner (ie the Y chromosome) is passed on generation after
generation, father to son, with little alteration. If two males share
"the same" Y chromosome the only way that is possible is if they got
it from a common male ancestor.

So how can WikiTree use the data? Most surname groups provide the
data, usually in excel spreadsheet format. Most genetic testing houses
also have "sift and sort" projects that look for matches without
reference to the surname. I am not sure of any cooperation between
testing houses but would suspect there is little.

What is needed is a fairly simple but potentially quite large database
into which people could enter their results in a standard excel spread
sheet format. Nothing too hard with that, even ensuring anonymity and
security. What is then needed is a fairly simple "sift and sort"
routine that takes your results and gives you those that match at the
67, 66, 65 and maybe 64 gene level. Beyond that pushes the probable
common ancestor back beyond the paper record.

Just as you now upload your gedcom and look for matches you would
upload your test results and again the system would tell you of any
matches. Then the hard work begins but at least it is targeted.

[Peter G]

I guess this raises the quesion of wether WikiTree is a viable format for this kind of project.

 

Would WikiTree end up with hundreds of pages that are not easily findable of DNA results or family projects? 

 

Or would we end up with 10s of thousands of pages of results that are mostly in different formats / layouts and not easily sortable?

 

Peter

 

---

From: John Hunter jh99...@gmail.com

[John Hunter]

At the moment you load a gedcom which goes into a database from which the web pages are produced. Most web sites are generated this way today. Now you can do a search which looks through the rest of the database to look for matches and it returns a page through which you can contact the profile owner whose data you match.

What I am proposing is to add the option of uploading another file which has your 67 marker test results, anything less than that is a waste of time. This would simply be another table in the database and could well be used to just produce another web page showing your results but I don't see any solid research need for that.

Again you could request a search and the system sifts through the results it has in the database and tells you of any close matches you have from 67 down to 64, a task best left to computers. The system then provides you with contact details as happens now. This test would put the common ancestor somewhere within the period of available records and available surnames.

A match on all 67 markers puts the probability of a common ancestor within 300 years at close to 100% and that is really what you want to know. The fewer matches you have simply reduces the probability at the 300 year mark or pushes the date back further than 300 years for a 100% probability of sharing a common ancestor within that time period (you can ignore that last sentence if you want, the important one is the one before it).

Note the genetic records need not have much to do with surnames as there are many reasons why a son can be raised with a name different to his biological father's name and that is one of the points for genetic testing. A third of my surname group doesn't have the surname of the group.

 

[John Hunter]

In retrospect I need to clarify something about the structure of the
data required. The assumption is that WikiTree is built around a
relational database; it has been a while since I have been near that
"stuff" so I could be entirely wrong, if so ignore this message
altogether.

My guess is that what would be needed is a table with rows and columns
(using layman's terminology) where each row would contain the DNA
results for each profile that has one. The number of columns in the
table would need to be something like 70 or so, one being the profile
identifier and 67 being the results for each gene. The result is
simply a number.

To recreate a web page from that is simple and the result could be an
additional page attached to each profile.

Once there are a number of records in the table a reasonably simple
query could sort and sift the data and find matches and close matches.
If a match is found then existing routines in WikiTree could take over
and open communication channels between the relevant profile owners.

From what I have seen of the DNA data from different testing centres
it is the same both in the genes that are tested and in the order of
presentation of the results. That may not be the case and would need
to be checked. Also a set of rules would need to be developed about
the layout of the file that would be used to upload the data. A simple
csv file should be enough.

Technically I don't see any problem with getting the data into the
system, testing it and notifying the relevant parties if a match is
found.

Why do it?

Simply because there are several testing laboratories and there is
little sharing of data between them. Each seem to carry out broad
match testing (ie across surnames and across geographical areas) but
there seems to be no cooperation between them. Some leakage does
occur, for example in my own surname group results are collected from
a variety of testing laboratories which extends the reach of the
surname group. However the surname is a construct and is an impediment
to the advancement of the research. As I have written previously my
own surname group has many different surnames in it and perhaps a
third have one other surname in particular. The surname is not what
unites us but the DNA record does.

So the reason to include this record in the WikiTree database is to
enable to collection of data in a central point irrespective of the
source of the data.
The yDNA test is only for males and enables the discovery of common
male ancestors within a time period that currently coincides with the
available paper record in many countries. Although the 111 test is
available its use is restricted to differentiating between those that
have an exact match at the 67 marker level. Anything less than 67
would not provide much value in my opinion so the initial results
entered should be restricted to the 67 test results.

The purpose of the test is to find a match or a very close match
irrespective of the surname. In my own case there was an exact match
which later was resolved to be a common ancestor about 300 years ago.
It seems that we had descended from two brothers meaning our common
ancestor was their father. More dramatically was the discovery of a
close match which turned out to be a third brother and the marker that
had mutated was one that was known to mutate within the time period
involved. The paper search that resulted turned up a diary that named
not three but four brothers and only a few months ago a living male
descendant of the fourth brother has been found and he has agreed to
undergo the DNA test so we await those results.

Similarly there is no point at this time in including the mtDNA test
results which is the test that is used for women. That DNA does not
have the variation required to be of much use to genealogists.

[Chris Whitten]

Thank you so much for this, John.

Your assumption is correct: We have a relational database (MySQL) that
could be used.

So the challenge wouldn't be technical. The user-interface would take
more time and effort to develop than the actual database. But the real
challenge, I think, would be getting people to use it.

If a half-dozen people upload their data, what good does it do? For
something like this, you probably need to have results for tens of
thousands of people in the database before you start seeing meaningful
matches. Maybe even hundreds of thousands, right?

Or maybe this would be useful for small targeted groups of people. If
you have a One-Name Study group who decided to use WikiTree to store
and sort their DNA data, maybe it would be useful for them?

Chris

> --
> You received this message because you are subscribed to the Google
> Groups "DNA and Wiki Genealogy" group associated with http://www.WikiTree.com
> To post to this group, send email to
> dna-and-wik...@googlegroups.com
> To unsubscribe from this group, send email to
> dna-and-wiki-gene...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/dna-and-wiki-genealogy?hl=en
>

--

[Peter G]

What would it take for Y-DNA results to be automatically propagated?  For example I upload my Y-DNA results. It would automatically create a link between my results to my male ancestors and to the male descendents of such ancestors. 

 

Still, the system is only really useful if you can get a lot of buy-in.

 

Peter

---

From: Chris Whitten <cwhi...@gmail.com>

To: dna-and-wik...@googlegroups.com
Sent: Sat, June 25, 2011 8:52:07 AM
Subject: Re: [Wiki DNA] Re: Test results on WikiTree

> dna-and-wiki-genealogy+unsub...@googlegroups.com

> For more options, visit this group at
> http://groups.google.com/group/dna-and-wiki-genealogy?hl=en
>



--
Chris Whitten
Creator of WikiTree.com
http://www.wikitree.com/wiki/Whitten-1

--
You received this message because you are subscribed to the Google
Groups "DNA and Wiki Genealogy" group associated with http://www.WikiTree.com
To post to this group, send email to
dna-and-wik...@googlegroups.com
To unsubscribe from this group, send email to

dna-and-wiki-genealogy+unsub...@googlegroups.com

[John Hunter]

Peter,

That is an extremely good point. Your DNA results should attach in some way to the profiles of all of your male line ancestors because in all likelihood it will be their DNA profile as well. If it is not then that is an important distinction to bring to light anyway. Similarly that result should be the DNA profile of all of the males that have descended from your oldest male line ancestor so potentially that result could have a spread over some hundreds or even thousands of males.

The other point you make is also very important. My testing laboratory has the results for approximately 7000 males (I think) which may translate to perhaps 1400 results at the 67 marker level so. Compared to the estimated 1 million males in WikiTree at the moment, it seems to be a minute proportion. However, as outlined in the previous paragraph, those 1400 results probably relate to perhaps 140,000 males - give or take - which becomes a significant number. None the less greater value comes from greater numbers so it is in some way incumbent upon the true believers to work to increase the numbers.

That gives rise to a number of opportunities. On one front WikiTree could work directly with the testing laboratories or at least with the coordinators of the various surname projects to get their clients/members to use WikiTree with obvious advantages for both parties and their constituents. It would also give the laboratories another reason to try to approach their clients to upgrade their results if they don't have test results at the 67 marker level. Most offer discounts for upgrades from time to time so this could just be another one. Maybe one of the laboratories could be persuaded to assist with the technical issues associated with data entry and, once the appropriate releases have been signed, the data could be entered directly from the laboratory.

Also WikiAnswers could be used to assist with spreading the understanding of what the value of DNA testing is for genealogists. Again many surname coordinators as well as laboratories have developed extensive web sites for their members/clients and some of those could be approached to participate in the broader project. Again it is a matter of putting relevant information centrally and making it freely available and searchable.

There are some questions already on WikiAnswers but their answers don't have the necessary depth or focus. I have placed a seed question "What is the value of DNA testing for genealogists?" under Society/Genealogy but perhaps a DNA sub-category would be more useful. As time permits I may get to a seed answer soon as well but anyone else can start the process.

John H

What is the value of DNA testing for genealogists?

Read more: http://wiki.answers.com/Q/Special:watchlist#ixzz1QQhMdD8m

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

[Chris Whitten]

Thank you, Peter and John! Now this conversation is really getting
interesting. This is exactly what I had hoped would emerge from this
discussion.

At first, we were talking about matching DNA test results.

Now, I think we're also talking about finding people who don't match.
This is potentially much more exciting.

Here's what I'm picturing. Check my logic on this.

You upload your DNA test results to WikiTree. If there are any exact
matches with other WikiTree users who uploaded their DNA test results,
you're alerted. If the other guy's family tree doesn't already overlap
with yours, you can get started figuring out what's missing or
incorrect.

But that wouldn't be the end of it. Peter and John talked about your
Y-DNA results propagating up your direct male ancestry on WikiTree
(son to father, father to grandfather, etc.) and then back down
through all your last male ancestor's male descendants (from
grandfather to father and uncles, from father and uncles to brother
and male cousins, etc.). As John pointed out, this could mean
attaching your results to hundreds or even thousands of profiles.

If each set of test results gets attached to hundreds of profiles, the
chances of two sets of results getting attached to the same profile
becomes very high.

Of course, many of the results attached to the same profiles would not
match each other. If they did match, the two people who uploaded them
would already know about it. Instead, we'll have results that should
match, but don't. This will lead to great opportunities for
investigation, and sometimes a clear indication that you need a
certain male cousin to get tested and upload his results.

It will probably make sense to somehow indicate on WikiTree how many
generational steps it took when we attached the DNA results to a
profile. With every step, a mistake in the genealogy becomes more
likely. Even your father might not be who you think he is. So, showing
the distance would help people figure out where the problems lie. Then
you can start looking at the sources for the genealogy.

There are so many opportunities here. Maybe someday you could even use
this to test tentative family relationships. For example, you might be
able to experiment to see what would happen if you changed this
father, or that son. How would that change propagate through the
worldwide family tree? Where would the DNA conflicts be?

Am I wrong, or is this really exciting?

Chris

[AJAskey]

Chris has asked me to put in my 2 cents here so here it comes...

Background:

I have had my yDNA and mtDNA tested at three companies. All three
returned the same results and I am now that confident I know some
history on both my paternal and maternal sides. My father and I are
U152+ which places our direct line male ancestors as far back as
10,000 years. Our paper trail begins in Edinburgh, Midlothian,
Scotland (circa 1691). The U152+ DNA haplogroup has been found in the
eastern side of Scotland. Prior to moving to the Scotland area
several thousand years ago, my ancestors lived in northern Italy and
Switzerland. Now if I find any information contrary to my U152+
footprint I can ignore it. This is a great help as records become
scarce in early times.

I met a 6th cousin from a tree that branched in the late 1700s.
Without DNA I would never have met this cousin and would not have
really known for sure if I was related where my paper trail led. Now
there is an almost certainty that the family history we both learned
independently as kids is correct. Without DNA testing there would be
no real confirmation of the paper trail. I have also met a 5th, 4th,
and 3rd cousin via DNA comparison. Each has provide information for
the trees that would not have been available otherwise.

Question:

Does standard genealogy strive to find the true genetic trees or is
geared toward finding families as they actually lived - meaning to
exclude non-paternal events (NPE)?

Suggestions:

If the users of WikiTree are more interested with family histories vs.
bloodline history, then I would exclude all DNA work. It will only
cause problems and my guess is will invalidate many trees beyond
100-200 years.

If users of WikiTree are interested in bloodline or a combination of
bloodline and family history then I would add DNA tags. I would
include them in some form of the following:

- Haplogroups - These are available to both males and females. I
suggest using the more recent SNP mutations(s) to add as attributes to
each tree. These will be searchable and will not mean anything to
anyone who does not care about DNA. They are short text fields such
as “U152” or “U5a1”. I can discuss these in another post if anyone is
interested. The short story is that if I am U152 positive, then my
paternal line is also U152 positive (back 1000s of years). Users can
attach multiple SNP designations but only the last positive one in the
phylotree is necessary.

- yDNA markers - It may be best to put a field for the ySearch.org ID
and then users can use that tool instead or recreating it on
WikiTree. Note that if two males are not in the same haplogroup then
they will not match on their yDNA markers.

- mtDNA - I suggest the same for mitosearch.org that I did for
ySearch.org. Only the mitosearch.org ID should be attached to a
person.
Propagating haplogroups - I suggest only doing this “up” the tree.
All of my male ancestors will have my U152 SNP. These can be
propagated down other male branches from the top also. I suggest
marking the original with a note showing who took the test and all
propagate tags should be marked with “derived from []”. Tested
haplogroups from others in the tree should override any derived
values. Any conflicts will need to be worked out by the owners of the
trees (and suggest a NPE occurred at some point).

I will be glad to address and issues or questions as they arise.

Thanks.
Andy Askey

[Andy Askey]

Chris - I would like to comment of your post from June.

> Now, I think we're also talking about finding people who don't match.
> This is potentially much more exciting.

And just as useful as matches. BUT, it will also cause disputes
against the DNA data as someone will always have a hard time giving up
on hours/weeks/months/years or research when it turns out that
prominent cousin has an unexpected father.

> You upload your DNA test results to WikiTree. If there are any exact
> matches with other WikiTree users who uploaded their DNA test results,
> you're alerted. If the other guy's family tree doesn't already overlap
> with yours, you can get started figuring out what's missing or
> incorrect.

The yDNA results data structures will need some thought as there can
be 111 individual markers and there will need to be a standardized
entry method and storage structure.

Also, some people test 25 markers while others test 67. The search
will need to compare the common markers only.

The mtDNA is similar but has many fewer markers. I doubt you want to
add the coding region which contain medical info and most will not
make this public.

>
> But that wouldn't be the end of it. Peter and John talked about your
> Y-DNA results propagating up your direct male ancestry on WikiTree
> (son to father, father to grandfather, etc.) and then back down
> through all your last male ancestor's male descendants (from
> grandfather to father and uncles, from father and uncles to brother
> and male cousins, etc.). As John pointed out, this could mean
> attaching your results to hundreds or even thousands of profiles.

Haplogroups (which are SNPs related to a geographical region) can
propagate up and down the tree. Haplotypes (y markers) should only
propagate upward. The reason is that some markers are picked by the
testing companies because they mutate at a high rate (every few
generations). For example, my 6th cousin and I match all markers
except 2. This is useful as we will be able to distinguish to which
branch a new match will belongs.

> It will probably make sense to somehow indicate on WikiTree how many
> generational steps it took when we attached the DNA results to a
> profile. With every step, a mistake in the genealogy becomes more
> likely. Even your father might not be who you think he is. So, showing
> the distance would help people figure out where the problems lie. Then
> you can start looking at the sources for the genealogy.

I use manual triangulation to determine tree accuracy. I match my
cousin and our common ancestor was born in 1742. I have high
confidence the trees main events are correct on our two branches.
Some sites have built triangulation into an algorithm and maybe you
can investigate that as a longer term project.

Andy